This project will develop a computer system for the automated analysis of sleep data including EEG, EOG, EMG, and respiration signals. The system will include a data base management system for maintaining a library of processed records from which one can select records sorted for a particular parameter or parameters such as age. The work will incorporate recent advances in Artificial Intelligence on the design of Expert Systems. The system will include software which allows the user to communicate with the computer in human language form. Much of the knowledge and expertise required for the operation of automated sleep analysis systems will be programmed into the computer. The system will also serve as a repository for knowledge accumulated by many sleep specialists. The computer will enable one to obtain a more detailed description of the sleep process than is currently obtained using visual scoring, and it will provide objective data which can be shared by all laboratories. In addition, the system will provide objective data which can be shared by all laboratories. In addition, the system will provide an economical method for providing sleep analysis to a large subject population. The system is to be evaluated with sleep records, a magnetic tape, which has been hand scored by three independent laboratories. In addition to EEG/EOG/EMG data the system will analyze respiration patterns, oxygen saturation levels, heart rate and leg myoclonus. The study will assist the establishment of objective criteria for the diagnosis and treatment of sleep disorders. The resulting system can provide a method for making sleep studies easily and economically available to a broad subject population.